A new delirium scoring system has shown excellent correlation with clinical outcomes in hospitalized elderly patients, including length of stay, functional decline, and death, investigators report.
In both short and long form, the Confusion Assessment Methods-S (CAM-S) is designed to complement the existing CAM, Dr. Sharon Inouye and her colleagues reported in the April 14 issue of Annals of Internal Medicine (2014;160:526-33).
"The short form (5-minute completion and scoring time), which is based on the CAM diagnostic algorithm alone, is quicker and simpler to rate; however, the long form (10-minute completion and scoring time) provides a broader range of severity scores in delirium and no-delirium groups," wrote Dr. Inouye of the Institute for Aging Research, Boston, and her coauthors.
"Unlike the Delirium Rating Scale, a clinician rater is not required for the CAM-S. Instead, well-trained research assistants can reliably conduct the assessments," the researchers wrote.
Both the short-form and long-form CAM-S instruments were validated in a group of 919 patients aged 70 years or older, who were scheduled for major surgery. The cohort was drawn from two extant study groups: the ongoing SAGES (Successful Aging After Elective Surgery) study,and Project Recovery, which ran from 1995 to 1998. Delirium was first rated by the existing CAM, and then according to the two versions of CAM-S.
The short-form CAM-S rates patients on four features of the CAM: symptom fluctuation, inattention, disorganized thinking, and altered level of consciousness. The most severe score is a 7. The longer form is based on 10 features, which also include disorientation, memory impairment, perceptual disturbances, psychomotor agitation, and sleep-wake cycle disturbance. The most severe score is a 19.
The measures had excellent correlation with each other, and with several clinical outcomes, the investigators said.
Length of hospital stay increased with increasing delirium severity across both forms, with an adjusted mean stay of 6.5 days for no delirium to almost 13 days with high severity in the short form. In the long form, length of stay increased from about 6 days to 12 days.
Hospital costs also tracked severity, ranging from an adjusted mean of $5,100 for no delirium to $13,200 for severe delirium in the short form. A similar pattern emerged in the long form, ranging from $4,200 to $11,400.
Functional decline was also highly correlated with score. On the short form, it occurred in 36%-68% of patients, depending on severity. In the long form, the range was 25%-61%. Cognitive decline showed a similar pattern.
In the short form, the cumulative adjusted rates of death within 90 days ranged from 7% to 27%, depending on severity. In the long form, the range was 7%-22%.
In the composite outcome of death or nursing home placement, results on the short form ranged from 15% to 51%, depending on severity. In the long form, the range was 13%-48%
While the Project Recovery data are more than 16 years old, the researchers said that this time lapse is not an issue because their primary interest is in comparison of outcomes among severity groups, which minimizes the importance of the internal values.
Also, "there may be inherent dependencies between CAM-S score and adverse outcomes," investigators wrote. "For example, patients with longer lengths of stay may have had higher CAM-S scores because of more opportunities for measurement."
The CAM-S score requires validation in groups younger than the age 70-plus patients addressed in the current study, researchers noted.
The National Institute on Aging funded the study. None of the authors reported having any financial declarations.
On Twitter @alz_gal